Lea Sztuk Haahr, the CESSDA Coordinator at the Danish National Archives, describes the main outcomes of the Danish project FAIR Across and how it helped researchers to understand and implement the FAIR Data principles.

The FAIR Across project is funded by DeiC (Danish E-Infrastructure Consortium) and managed by the Danish Forum for Data management.

The project is a bottom-up project that started in March 2018 and ran until the end of the year. It started with informal discussions between several universities and the Danish National Archives (DNA) and a shared view that the FAIR data principles should be addressed in a more practical way.

The group was led by the Danish National Archives (DNA) and included representatives of Aalborg University (AAU), the Royal Danish Library (KB), Copenhagen Business School (CBS), Copenhagen University (KU), and the Danish Technical University (DTU).

The project was expected to deliver material that could help researchers understand FAIR and implement changes in order to make their research more FAIR. We of course planned the usual deliveries: posters, postcards and a toolbox related to the research process. However, we also saw a need for delivering something different – something people would remember. That’s how the FAIRy tale was born, a tale that could demonstrate how to put theory into praxis. It also meant that we could show the reader, in a light-hearted way, what happens when data do not have metadata or a DOI.

The project kicked off with two days where we looked very closely into the FAIR principles and allowed ourselves to open them up, and related them to the different disciplines we represented. We often broke off into small groups, discussing small elements of the principles. At the end of these two days we realised that we wanted the project to deliver concrete material based on real life researcher problems.

All of the principles were not necessarily fully applicable to all research disciplines, and this we wanted to look into. We also wanted to make all project deliverables with the specific approach: “What’s in it for me?”. As far as we were concerned, researchers should not be imposed more work unless they can see a tangible benefit.

The deliverables of these two days were some myth buster postcards (example) and a template for interviewing researchers. We mainly wanted to find out whether they had prior knowledge about the FAIR data principles. Researchers that we would interview would have to pick a specific research project and describe where they were in the research process.

We then considered a range of problems related to their current research process and to data in general, such as:

• How data is stored
• Use (or no use) of metadata
• Change in staff
• What will happen to data when the project ends
• Etc.

All the project participants represented very different research disciplines, and it was therefore part of the project that each participant conducted several interviews with researchers representing these disciplines. My interviews reflected the researchers in touch with the Danish National Archives – social sciences data and sensitive heath data. Researchers were contacted and interviews were conducted during June 2018.

The project participants represented very different research disciplines:

• AAU: Biotechnology
• DNA: Sensitive Health Data
• DNA: Not sensitive Social Science Data
• KU: Linguistic Data
• KU/KB: Language and Literature
• DTU: Wind Energy
• CBS: Social Science
• CBS: Social Science (video data)

When the interviews were done, all the data were gathered in schemas and templates making sure that the information was comparable across disciplines and across different stages of research processes. This also revealed that the knowledge about FAIR varied across the disciplines.

Based on the data we identified the general knowledge of FAIR in the various communities. We also identified the most common problems and related these to different steps in the research process.

The next step was to present some solutions, so we could convince the researchers interviewed to be more aware and in line with the FAIR data principles. We therefore hosted a “Train-the-Trainer” arrangement, where international experts were invited to present different tools that could improve the level of compliance with the FAIR data principles. We focused on tools and services, which were then assessed according to the FAIR data principles.

After the “Train-the-Trainer” event, the project group had the tools that it needed to solve some of the problems researchers had mentioned during the first interviews. The tools were ranked and connected to different steps in the research process. This activity led to the creation of a toolbox. It allows a researcher to see specific tools that solve specific problems, and at the same time improving the “FAIRness” of the data.

The tools are classified by their availability (national or international), discipline, FAIR dimensions, phases in the research life cycle. They are then named, described and a link is provided to the tool.

The next step was to verify these and test them during a second round of interviews. We all conducted a second round of interviews with the same researchers and presented the tools mentioned before. Again we all had the same interview guide and template to record the information from these interviews. In general, the interviewed researchers were very satisfied with the toolbox. The toolbox offered them “real solutions to real problems”, when trying to improve their level of FAIRness of the data.

Data from the interviews were again mapped and compared, and the toolbox was verified so it was ready for use by researchers and others with interest in the FAIR principles.

The fairy tale was also written in this period. We were three authors behind the project, Karsten Kryger Hansen from Aalborg University, Mareike Buss from Copenhagen Business School and me. It was quite a challenge to turn all of the principles into praxis and then into a fairy tale. We could also see that the fairy tale concept was unique and could carry the whole project and our efforts therefore seemed worthwhile. The writing process was harder than expected, especially as we were located in three different regions of Denmark, but extremely rewarding in many aspects.

On 20 November 2018, the closing conference of the project took place in Copenhagen. The researchers that had been actively engaged in the project took part and shared their views about what it was like to be a part of the project and what they had learned about FAIR data. We also had a presentation from CESSDA (Ulf Jakobsson from SND) where the CESSDA Data Management Expert Guide was presented (presentation).

The project really made a great impact for both the project participants and the researchers who volunteered. Valuable professional networks between data experts and researchers were created, new projects are emerging and there is much more interest in FAIR Data in the different research communities. Support from libraries, research institutions and archives for FAIR Data is higher than ever, and knowledge is definitely improving all the time. In concrete terms, the toolbox developed by the project now supports researchers and the expert community is available for giving advice.

We would like to build on this project and the knowledge and tools gathered during the process and are planning next steps. The challenge as we see it is not only promoting the use of the toolbox, it is also about changing the culture around how data is handled throughout the data life cycle.

The FAIRy tale was published in December 2018 and holding the actual FAIRy tale book in my hands was an amazing feeling. Everyone involved in the project has helped to spread the word and the FAIRy tale has definitely had the impact that we had hoped for. We have been contacted by many institutions all over Europe wishing to publish the FAIRy tale on their websites. Researchers are also very impressed – the fairy tale concept makes FAIR understandable and relatable.

More information:

• Visit the Escience vidensportal: https://vidensportal.deic.dk/en
• For specific information on FAIR data in Danish and in English.
• Download the FAIRy tale and the poster.
• Contact: Lea Sztuk Haahr, CESSDA Coordinator, The Danish National Archives, lsh@sa.dk.

The Danish CESSDA Service Provider: As a part of the Danish National Archives, we are dedicated to the dissemination of data – be it survey data or public registers – for scientific purposes.